I have collected data from a person performing the same action for 10 iterations. As shown in the figure, signals from all iterations look very similar, but contain some unwanted noise in some occasions. Here, I want to identify the starting point of each movement reliably. For now I use a threshold on the 1st derivative of the data, however it detects it wrongly in some iterations (No 8 in the figure).
Is there a method to identify the starting point of the signal, considering the similarity of the signals? For example, I was thinking of using 1 signal as a baseline, and trying to detect the similar features on other signals to identify the starting point (using machine learning). However, it is only an idea and I have no clear way of doing it.
What would be the best way to do it (and how)? I am open to ideas.